Ozonolysis of cis- and trans-diisopropylethylene in the presence of

Jul 9, 1970 - Ozonolysis of cis- and irans-diisopropylethylene in the presence of lsO-labeled isobutyraldehyde leads to in- corporation of the lsO lab...
0 downloads 0 Views 518KB Size
J. Org. Chem., Vol. $6, No. 8, 1971

Cis- AND h73%3-DIISOPROPYLETHYLENE

1103

Ozonolysis of cis- and trans-Diisopropylethylene in the Presence of 180-LabeledIsobutyraldehyde R. W. MURRAY" AND R. HAGEN' University of Missouri-8t. Louis, St. Louis, Missouri 65121 Received July 9, 1970 Ozonolysis of cis- and trans-diisopropylethylene in the presence of 180-labeled isobutyraldehyde leads to incorporation of the 1 8 0 label in both the ether and peroxide bridges of the resulting diisopropylozonide. More of the 1 8 0 label is incorporated in the peroxide bridge at lower temperatures.

We have recently shown2that ozonolysis temperature has an important effect on ozonide isomer ratio and yield. These results are important to the problem of the mechanism of ozonolysis, since they contain support for an additional mechanism for the reaction which we had suggested e a r l i e ~ - . ~According ,~ to our earlier suggestion, under suitable reaction conditions the initial olefin-ozone adduct 3 (Scheme I) might react directly with the aldehyde to give 10 and eventually ozonide 13 with an isomer ratio which differs from that produced in the accompanying Criegee6 zwitterion pathway 3 -P 8 -P 14.

The use of lPO-labeled aldehyde in principle permits one to distinguish the two reaction pathways described. Thus the pathway through a partially cleaved initial adduct 6 eventually places the l80only in the peroxide] whereas the Criegee pathway places the l80only in the ether bridge of the final ozonide, 14. I n some early works it was found that, in the case of trans-diisopropylethylene ozonized a t low temperature in the presence of ls0-labeled acetaldehyde, a sizable amount of the l80label is found in the peroxide bridge of the methylisopropylozonide product. In this case the location of the label was determined by chemical means. FlisziLr and coworker^^-^ have also used the l80labeling technique and have concluded that, in the ozonolysis of phenylethylenes in the presence of l8O-labeled benzaldehyde, the l80label is incorporated exclusively in the ether bridge. I n this work, mass spectrometry was used to determine the location of the label. It should also be noted that most of these ozonolyses were carried out at room t e m p e r a t ~ r e . ~These workers also felt that the simple Criegee pathway was inadequatelo and suggested an alternative path to ozonide formation, 3 --t 2 + 11 -P 14, which places the l80label in the ether bridge. For the sake of completeness, Scheme I contains all possible structures 2, 3, 4, and 5 for the initial adduct. I n fact, structure 3 is now considered the preferred structure except in those cases of hindered 1 olefins in which 5 seems to be the most likely structure.ll (1) Ciba Photochemical, Ltd., Fribourg, Switzerland. The work was carried out at Bell Telephone Laboratories, Inc., Murry Hill, N. J. (2) R. W. Murray and R. Hagen, J . Ovg. Chem., 86, 1098 (1971). (3) R . W. Murray, R. D. Youssefyeh, and P. R. Story, J . Amer. Chem. Soc., 89, 3143 (1988). (4) R. W. Murray, R. D. Youssefyeh, and P. R. Story, ibid., 89, 2429 (1987). (5) R. Criegee, Rec. Chem. Progr., 18, 111 (1957). ( 8 ) P. R. Story, C. E. Bishop, J. R . Burgess, R. W. Murray, and R. D. Youssefyeh, J . Amer. Chem. Soc., 90, 1907 (1988). (7) 8. FlisaBr, J. Carles, and J. Renard, ibid., 90, 1384 (1988). (8) S. Flisz&r and J. Carles, ibid., 91, 2837 (1989). (9) J. Castonguay, M. Bertrand, J. Carles, 8. Fliszhr, and Y. Rousseau, Can. J . Chem., 47, 919 (1989). (10) 8.Flisz&rand M . Granger, J . Amer. Chem. Soc., 92, 3381 (1970).

Scheme I also includes an alternative form 7 of a partially cleaved initial adduct as well as indicating that the zwitterion pathway may include both cage and solvated possibilities. Results and Discussion The results obtained from mass spectrometric analysis of diisopropyl ozonides produced by ozonolysis of 0.2 M solutions of cis- and trans-diisopropylethylene in the presence of varying concentrations of 180-labeled isobutyraldehyde are given in Table I. It should be noted that separate results were obtained for the cis and trans ozonide in order to obtain the maximum amount of information regarding the total mechanism. The total labeling was obtained from the molecular ion peaks 160 and 162. The labeling of the ozonides in the ether bridge was derived from the peaks a t m/e 128 and 130, assuming the loss of two oxygen atoms (mass 32 or 34) occurs only from the peroxide bridge. I n this connection, it is important to note that loss of methyl groups is not important, leading to small peaks at M - 15 and 14 - 30. The errors reported in Table I are calculated for the 90% confidence limit. The 180-labeling was calculated as follows. H (M

+ 2)

Hn (M

+ 2)

The index zero refers to unlabeled ozonide and H corresponds to the peak heights at the given m/e ratio. The labeling of the ozonides in the peroxide bridge is calculated as the difference between the total 180-labeling and the labeling in the ether bridge. As shown in Table I, considerable amounts of lSOlabeled aldehyde are incorporated into the ozonide at both ends of the temperature range of interest. Furthermore, l80is definitely incorporated into the peroxide bridge and more l80is incorporated into the peroxide bridge a t the lower temperatures for both the cis and trans olefins. These results are thus compatible with the conclusions reached earliers in a similar system where the l8O distribution was determined by chemical methods. Two significant differences distinguish the FliszAr, et al., and that reported here. All of the olefins studied by Flisz&rl et al., contain an aromatic substituent a t the double bond. I n these cases, the phenyl group would be expected to stabilize the zwitterion thus favoring this reaction pathway. Second, and perhaps more important, the lowest temperature studied by these workers was - 7 P with most of the ozonolysis to (11) R. W. Murray, Accounts Chem. Res., 1, 313 (1988).

1104 J. Org. Chem., Vol. 56, No. 8, 1971

MURRAY AND HAQMN SCHEME I

-

+

,0--0 0

RCH-CHR

/o 00 I

R'CHO*

/

RCH-CHR

\

R'CHOI

v

\ + RCH-CHR

1

13

10

0/O-.O-

0

A

W

R'CH-CHR

6

2

RCH-CHR

%

0-0

RCH-CHR

2

I

R'CHO*

I

P-O,,CHR'

RCH,

-RCHO

RCH

~- H R

R$H

\

11

\ *O-puD'

7

Or 14

1:

one-bond cleavage intermediates

seven-membered ring intermediates

J.

SR'CHO"

I

RfH solvent

i

/

/

9a

RCHO solvent 9b two-bond cleavage intermediates TABLEI MASSSPECTROMETRIC RESULTS FOR I~O-LABELED DIISOPROPYL OZONIDES~~~

Olefin

Aldehyde Ozonizaconcn, tion temp, mol/l. OC Warm-up

-Total 180 labeling, %-Cis ozonide Trans ozonide

180 labeling in the -ether bridge, %Cis ozonide Trans ozonide

1 8 0 labeling in the -peroxide bridge, %Cis ozonide Trans ozonide

11.0 f 1.8 12.2 f 1.0 0.4 f 2.1 6.6 f 1.6 -117 Slow 11.4 f 1.1 18.8 i 1.4 14.3 f 3.3 4.0 f 4.5 11.6 f 3.6 Fast 22.2 f 0.8 18.2 f 4.4 -117 25.9 f 1.4 3.8 f 1.2 3.6 f 2.0 22.2 f 1.9 23.0 f 0.8 -1 Fast 19.2 f 0.9 25.8 f 0.8 5.8 f 2.6 12.7 f 1.8 11.3 f 0.7 1.0 -122 Fast 22.0 f 0.3 24.0 f 1.6 16.2 f 2.6 0.2 -118 Slow 18.8 f 1.4 16.9 f 0.6 15.3 f 1.2 1.7 f 0.7 3.5 f 1.8 15.2 f 0.9 4.2 f 1.9 10.3 =k 1.8 6.8 f 1.5 Fast 17.8 f 1.5 13.6 f 1.3 -113 17.1 f 1.0 4.6 f 1.9 3.8 f 2.7 16.4 f 2.6 15.6 f 1.4 -1 Fast 20.2 f 1.3 20.2 f 0.8 7.6 f 1.6 0.1 f 2.4 10.7 f 1.6 17.2 f 1.6 0.4 -120 Fast 17.3 f 1.8 18.3 =k 0.4 a Obtained by ozonolysis of 0.2 M solutions of cis- and trans-diisopropylethylene in the presence of 180-labeled isobutyraldehyde. Errors are given for the 90% confidence limit. * 1 8 0 enrichment of the aldehyde was 34% for the experiments at 0.2 M aldehyde concentration and 22% in the other cases.

2

0.2

X Y

give lsO-labeled ozonides being carried out a t room temperature. The mechanism we have ~ u g g e s t e d per,~~~ mitting ls0 incorporation in the peroxide bridge, gives a role to the initial adduct and thus would be expected to be more significant at lower temperatures. The results in Table I are a t least consistent with this expectation. It may be significant that Flisehr, et u Z . , ~ considered that up to 10% l80incorporation in the peroxide bridge at the lowest temperature used (-78" for cisstilbene and -20" for trans-stilbene) would be consistent with their data. However, the stilbene ozonide obtained from the ozonolysis of styrene in the presence of 180-labeled benzaldehyde a t -78" was labeled exclu-

sively in the ether position. The extension of l80labeling studies to lower temperatures for olefins with aromatic substituents should perhaps be completed before a final conclusion concerning the mechanism is made for these cases. In agreement with the results found in ref 2, solutions of the trans olefin ozonized at low temperature show less incorporation of bulk aldehyde on slow warm-up than on fast warm-up. For the trans olefin, also, fast warmup at low temperatures leads to a greater incorporation of l*O into the peroxide bridge than slow warm-up, with the trans ozonide receiving about two-thirds of the total. For the cis olefin, with its less stable initial adduct,

J. Org. Chern., VoZ. 36,No. 8, 1871

cis- AND trans-DIISOPROPYLETHYLENE RANGES FOR

Olefin

TABLE I1 LABELING IN THE ETHER BRIDGE (IN PERCENTOF THE TOTAL LABELING) AND OZONIDECIS/TRANSRATIOSFOR LABELING IN THE ETHER BRIDGEA N D LABELING IN THE PEROXIDE BRIDGE.

THE 1 8 0

FOR THE

Aldehyde oonon, mol/].

Ozonization temp, "C

Warm-up

-117 Slow -117 Fast -1 Fast -122 Fast 1.0 0.2 -118 Slow -113 Fast -1 Fast 0.4 -120 Fast a Obtained from ozonolyses of 0.2 M solutions of enriched in l80.

2

0.2

W

Most probable ranges of the 1 8 0 labeling in the ether bridge in % ---of total labeling-Cis ozonide Trans ozonide

SCHEME I1 0-0-

15

--Most

probable ranges of the ozonide cis/trans ratiosOzonide labeled in Ozonide labeled in Total ozonide the ether bridge the peroxide bridge

74 to 100 52/48 55 to 76 43/57 to 55/45 0/100 to 35/65 60 to 100 34/66 40 to 72 29/71 to 51/49 0/100 to 35/65 77 to 91 76 to 96 35/65 29/71 to 35/65 20/80 to 63/37 61 to 87 41 to 54 44/56 47/53 to 58/42 15/85 to 38/62 70 to 95 74/26 6 to 15 94/6 to 98/2 23/77 to 51/49 64 to 91 29 to 52 73/27 80/20 to 88/12 34/66 to 66/34 66 to 90 66 to 98 56/44 49/51 to 61/39 35/65 to 88/12 82 to 100 32 to 51 70/30 80/20 to 88/12 0/100 to 39/61 cis- and trans-diisopropylethylene in pentane in the presence of isobutyraldehyde

both fast and slow warm-up a t low temperature lead to sizable amounts of '*O being incorporated into the peroxide bridge. Here again the trans ozonide receives a greater percentage of the total. The experiments a t 0.4 and 1.0 M aldehyde concentrations were carried out to investigate the importance of the different labelings at high aldehyde concentration. These data are important to an interpretation of the relative labeling in the ether and peroxide bridges since it is at least conceivable, although probably not very likely, that some lSOcould be incorporated into the peroxide bridge as shown in Scheme 11.

9a

1105

16

At an aldehyde concentration of 0.2 M , the label went to about two-thirds or more into the ether bridge of the cis ozonide for all reaction conditions. I n the trans ozonide, the label is also placed preferentially in the ether bridge in the ozonolyses at high temperatures. However, for the trans ozonide at low temperature, the labeling in the peroxide bridge becomes more important for both cis and trans olefins and, in the case of the cis olefin, even becomes dominant. From the ranges reported in Table I1 for the cis/trans ratios of the ozonide labeled in the ether bridge, it appears that the cis and trans olefins are giving different cis/trans cross ozonide ratios at all temperatures, the cis olefin giving a higher percentage of cis ozonide and both olefins tending toward a higher percentage of cis ozonide at lower temperatures. The ozonide with labeling in the peroxide bridge contains a larger percentage of trans ozonide in the case of the trans olefin while the cis olefin can give a higher percentage of cis ozonide. Summary and Conclusions

If this scheme were an important source of lSO labeling in the peroxide bridge, then it would be expected to play a reduced role at higher aldehyde concentrations where there is greater opportunity for interception of 9a prior to equilibration through 15 and 16. I n fact, the relative importance of peroxide and ether bridge labeling does not change greatly with aldehyde concentration, although one does observe a higher incorporation of bulk aldehyde. It is significant also that Fliszhr and coworkersl2 have demonstrated in an elegant way that structure 16 does not participate in the mechanism of ozonide formation in the ozonolysis of trans-4-methyl2-pentene. Table I1 gives the mass spectral data in a form which is perhaps more suitable for a discussion of the mechanisms leading to cross ozonide. It contains the labeling in the ether bridge as a percentage of the total labeling and estimated ranges for the ozonide cis/trans ratios for both ether and peroxide bridge labeled material. The ranges given are estimated to be the most probable values and were calculated from the data in Table I by taking the extreme cases of the 90% confidence limits. (12) G. Klutsoh, J . Grignon, J. Rensrd, and S. FliszBr, Can. J . Chem., 48, 1598 (1970).

It has been shown that ozonolysis of cis- and transdiisopropylethylene in the presence of 180-labeled isobutyraldehyde leads to incorporation of the l80label at both the ether and the peroxide bridge of the resulting disopropylozonide. The incorporation of the l*O label in the peroxide bridge is temperature dependent with more of this labeling occurring a t lower temperatures. These results are consistent with our earlier suggestion314that ozonide formation by the Criegee zwitterion pathway may be accompanied by an additional pathway involving reaction of the initial adduct 3 with aldehyde. I n the simplest interpretation, the per cent labeling in the ether bridge could be taken as the percentage contribution of the Criegee zwitterion pathway. The peroxide bridge labeling would then indicate the percentage contribution of some other pathway, possibly the aldehyde-initial adduct pathway. On this basis, for the trans olefin, the Criegee zwitterion pathway is most important (80%) a t - l o with fast warm-up. Its lowest contribution (60%) is a t -122", fast warm-up and high added aldehyde concentration. For the cis olefin, maximum influence of the zwitterion path is also observed a t -lo (79.5%) and fast warm-up, while the

1106

J. Org. Chew., Vol. $6, No. 8,1971

lowest contribution (60%) of this path was observed a t - 113' and fast warm-up. According to the working hypothesis suggested earlier by US,^^^ operation of the path 6 + 10 --+ 13 should give relatively more cis ozonide from the cis olefin than from the trans olefin. As yet there does not appear to be a way to put this prediction on an absolute basis. The results shown in Tables I and I1 indicate that if cis and trans olefins are compared under the same conditions, then in all cases studied the cis olefin gives relatively more cis ozonide in the nonzwitterion pathway. I n the other cases studied, which are not directly comparable, the trans olefin with 1.0 M added aldehyde and a t -122" gives more cis ozonide than the cis olefin a t 0.4 M added aldehyde a t - 120". In this last comparison, the cis olefin gives very little cis ozonide contrary t o the stereochemical predictions made earlier. 3 , 4 While the relative ozonide cis/trans ratios observed are consistent in most cases with the predictions previously madej314the differences are small and cannot be regarded as substantiating these predictions. In all cases except one ( - l o ) the trans ozonide incorporates more l80labeling than the cis ozonide. It is difficult to go from these data to an absolute ozonide cis/trans ratio for the nonzwitterion pathway. The most probable ranges are given in Table 11. These ranges do not permit any definite statement on the stereochemical course of the additional reaction pathway. Experimental Section The ozonolysis and analytical procedures have been described.2 Preparation of 180-Labeled 1sobutyraldehyde.-Isobutyraldehyde and 180-labeled water (YEDA, normalized in hydrogen) were mixed and kept a t room temperature for several days. The water was separated from the aldehyde on a 10 ft by l/1 in. column filled with 80-100 Poropak W (He flow rate 1-6 ml/sec, column temperature 135'). Minor impurities, such as acids, were removed from the aldehyde by means of a 30 ft by a / ~in. column containing 5Oj, XF-1150 on Chromosorb G (He flow rate 10 ml/sec, column temperature 57"). Two lots of different 1 8 0 labeling were prepared. Aldehyde, 347, enriched in l80,was used for the experiments with 0.2 M aldehyde concentration, whereas the experiments with 0.4 and 1.0 M aldehyde concentrations were carried out with isobutyraldehyde, 22% enriched in 1*0. The percentage enrichment was determined from mass spectra. Preparation of 180-Labeled 0zonides.-The 180-labeled diisopropylethylene ozonides were produced by ozonizing 0.2 M solutions of cis- or truns-diisopropylethylene in pentane in the presence of varying concentrations of 180-labeled isobutyraldehyde. Ozonolysis was carried out to 50% conversion in a procedure similar to that described for the unlabeled aldehyde experiments.

MURRAY AND HAGEN Mass Spectra of the Diisopropylethylene Ozonides.-The mass spectra were run on a Consolidated Electronics Corporation Model 21-104 mass spectrometer. Since the ozonides decompose a t an appreciable rate above ca. 80°, i t was necessary to vaporize the samples a t room temperature. The source temperature was 165"and the ion voltage was set a t 70 V. Under these conditions, the molecular ion peak was quite stong, but by no means the strongest signal. The most intense lines were a t m/e 43, 41, 72, and 56. Above 73 the most intense line was a t m/e 117 (M - 43), which corresponds to a fragment where the molecular ion has lost an isopropyl group. The intensities of the mass peaks in the region important for this investigation are given below (relative to the molecular peak intensity = 100). r---Relstive intensities of mass peaks (m/e)----162 161 160 159 130 129 128 127

Cisozonide 1 . 1 8 . 7 (100) 1 . 7 0.12 0.39 2.78 2.48 Transozonide 1 . 1 8 . 8 (100) 3 . 4 0.18 0.78 6.57 3.72 These intensilies were obtained as an average of several mass spectra traces of unlabeled diisopropylethylene ozonides. The peaks a t m/e 158 and 126 were negligible and the peaks a t m/e 145 (M - 15) and 144 (M - 16), which correspond to fragments produced from the molecular ion by loss of a methyl group or an oxygen atom, were generally much smaller than the peak a t m/e 128. I n some cases the peaks a t m/e 145 and 128 were of the same order of magnitude. The relative intensities of the peaks a t m/e 127 and 128 were rather characteristic for the two ozonide isomers although they varied occasionally for no obvious reason. The total l80labeling of the ozonides was calculated from the m/e peaks at 160 and 162 and the labeling in the ether bridge was determined from the peaks a t m/e 128 and 130, taking into account the relative peak intensities 1(162)/1(160) and 1(130)/1(128) obtained from the unlabeled ozonides. Generally two separate gpc collections from the same ozonolysis were analyzed with the mass spectrometer for each isomer with three traces taken for each collection. Since no obvious differences between two analogous collections were obtained, the average of all six traces was calculated and reported in Table I. Sometimes the amounts of ozonide available were so small that the two collections had to be combined for the mass spectral analysis and in these cases six traces were taken. The errors were calculated for the 90% confidence limits. They do not include possible errors due t o relative intensity variations of 1(162)/1(160) and 1(130)/1(128) for unlabeled samples. The labeling in the peroxide bridge was calculated by difference and assuming that the following formula applies.

Registry No. -cis-Diisopropylethylene, 10557-44-5 ; trans-diisopropylethylene, 692-70-6; 180-labeled isobutyraldehyde, 27720-65-6. Acknowledgment.-R. W. M. thanks the National Science Foundation for partial support of this work through Grant No. G P 10895.